Computer Science and Engineering, University at Buffalo
Spring Semester 2021
.red {color:red} .blue {color:blue} .gr {color:green;font-weight:bold} .highlights {background-color:#ef9c7f;padding:1em}- Varun Chandola (lead instructor; chandola[at]buffalo.edu)
- Deen Dayal Mohan (TA; dmohan[at]buffalo.edu)
- Seokmin Choi (TA; seokminc[at]buffalo.edu)
- Jie Zhang (TA; zhang326[at]buffalo.edu)
- Enshu Wang (TA; enshuwan[at]buffalo.edu)
Note
Students are strongly encouraged to use the Piazza's private messaging option to contact the intructors to ensure that the messages are dealt with promptly.
https://cse.buffalo.edu/~chandola/machinelearning.html
Every Monday, Wednesday and Friday - 1.50 to 2.40 PM, virtually on Zoom
Note
All lecture videos will be made available after the class. The links will be posted on Piazza and also available here. Please ensure that your video is turned off and the microphone is on mute. Use the zoom reactions and chat to interact with the instructor.
| Who? | When? | Where? |
|---|---|---|
| Chandola | Fridays 3.00 PM - 5.00 PM | Virtually on Zoom |
| Deen | Mondays 6.30 PM - 7.30 PM | Virtually on Zoom |
| Seokmin | Mondays 9.00 AM - 10.00 AM | Virtually on Zoom |
| Jie | Thursdays 2.30 PM - 3.30 PM | Virtually on Zoom |
| Enshu | Tuesdays 1:00 PM - 2:00 PM | Virtually on Zoom |
CSE 250 and (EAS 305 or MTH 411 or STA 301 or MTH 309).
Note
This course requires a strong background in linear algebra, advanced calculus and statistics. Please refer to the FAQs for more.
l
| Week | Topic | Pre-requisites |
|---|---|---|
| 1 | Introduction and Basics | |
| **Supervised | Learning::Linear Models** | |
| 1 | Linear Regression | Linear Algebra,Gradient Descent Optimization, Matrix Calculus |
| 2 | Logistic Regression/Perceptrons | Newton's Method |
| 2-3 | Support Vector Machines | Constrained Optimization, Lagrangian Methods |
| **Supervised | Learning::Non-linear Models** | |
| 4 | Non-linear Regression | |
| 4 | Regularization | |
| 5 | Kernel Regression | |
| 5 | Kernel Support Vector Machines | |
| 6-7 | Neural Networks | |
| **Statistica | l Learning** | |
| 8 | Generative Models | Laws of Probability, Statistical Distributions, Moments |
| 9 | Bayesian Learning Methods | Bayes Rule |
| 10 | Bayesian Classification | |
| 11 | Bayesian Linear Regression | |
| **Fairness a | nd Transparency Issues** | |
| 12 | Fairness in Machine Learning (PA3 Review) | |
| 12 | Interpretable Models (Decision Trees) | |
| **Unsupervis | ed Learning** | |
| 13 | Clustering (k-Means/Spectral) | Linear Algebra (Eigenvalue Decomposition) |
| 14 | Dimensionality Reduction Methods (Principal Component Analysis) |
| Deliverable | Release Date | Due Date |
|---|---|---|
| Gradiance 0 | Feb 1 |
|
| Gradiance 1 | Feb 10 |
|
| PA 1 | Feb 8 | Mar 5 |
| Gradiance 2 | Feb 17 |
|
| Gradiance 3 | Feb 24 |
|
| PA 2 | Mar 8 | Apr 9 |
| Gradiance 4 | Mar 3 |
|
| Gradiance 5 | Mar 10 |
|
| Gradiance 6 | Mar 17 |
|
| Gradiance 7 | Mar 24 |
|
| PA 3 | Apr 12 | May 7 |
| Gradiance 8 | Apr 1 |
|
| Gradiance 9 | Apr 8 |
|
| Gradiance 10 | Apr 15 |
|
| Gradiance 11 | Apr 22 |
|
| Gradiance 12 | Apr 29 |
|
Note
* Gradiance quizzes
- Will be released every Wednesday at 9.00 AM EST
- Due next Tuesday at 11.59 PM EST
- Gradiance 0 will not be evaluated (warm up)
- All assignments are electronically due on Fridays by 11.59 PM EST through UBLearns.
- Programming Assignment 1 - This assignment will focus on building linear models for supervised learning. This will include implementing a linear regression model for regression, and three classification models, viz., logistic regression, perceptron, and support vector machine (SVM).
- Programming Assignment 2 - In this assignment, your task will be to explore non-linear machine learning models to learn from text and image data.
- Programming Assignment 3 - This programming assignment has two parts. In the first part, you will implement a Naive Bayes Classifier and test it on a publicly available data set. In the second part, you will manipulate the data characteristics to understand how classifiers get impacted by the underlying bias in the training data. Focus will be on developing a COMPAS style risk assessment system.
- Kevin Murphy, Machine Learning: A Probabilistic Perspective, MIT Press, 2012.
- Tom Mitchell, Machine Learning. McGraw-Hill, 1997.
- Chris Bishop, Pattern Recognition and Machine Learning, Springer, 2006.
- David Mackay, Information Theory, Inference, and Learning Algorithms, Cambridge Press, 2003.
- Trevor Hastie, Robert Tibshirani and Jerome Friedman, The Elements of Statistical Learning. Springer, 2009.
- Richard S. Sutton and Andrew G. Bart, Reinforcement Learning: An Introduction. MIT Press, 2015.
- Short weekly quizzes using Gradiance (12) -- 20%
- Programming Assignments (3) -- 45%
- Mid-term Exam (virtual-UBLearns, open book/notes) -- 15%
- Final Exam (virtual-UBLearns, open book/notes) -- 20%
- Final grade (Tentative)
- A [92.5,100]
- A- [87.5,92.5)
- B+ [82.5,87.5)
- B [77.5,82.5)
- B- [72.5,77.5)
- C+ [67.5,72.5)
- C [62.5,67.5)
- C- [57.5,62.5)
- Mid-term Exam March 19, 1.50 PM - 2.40 PM, virtually using UBLearns
- Final Exam May 14, 11.45 AM - 2.45 PM, virtually using UBLearns
Note
The mid-term will held during the regular Friday lecture.
- Students are expected to act in a professional manner during the virtual classes and office hours.
- Programming assignments will be graded and returned to students.
- Late submission of assignments will receive a grade of zero.
- No late submission of Gradiance quizzes are allowed. The quizzes will automatically become unavailable immediately after the due date and no accomodations will be made for missed quizzes.
- Students are encouraged to discuss assignments and share ideas, but each student must independently write and submit their own solution.
- Makeup exams will be given in the following circumstances only:
- You contact the instructor prior to the exam
- You have a valid and documented reason to miss the exam
If you have a disability and may require some type of instructional and/or examination accommodation, please inform me early in the semester so that we can coordinate the accommodations you may need. If you have not already done so, please contact the Office of Accessibility Services (formerly the Office of Disability Services) University at Buffalo, 25 Capen Hall, Buffalo, NY 14260-1632; email: stu-accessibility@buffalo.edu Phone: 716-645-2608 (voice); 716-645-2616 (TTY); Fax: 716-645-3116; and on the web at http://www.buffalo.edu/accessibility/. All information and documentation is confidential. The University at Buffalo and the School of Engineering and Applied Sciences are committed to ensuring equal opportunity for persons with special needs to participate in and benefit from all of its programs, services and activities.
This course will operate with a zero-tolerance policy regarding cheating and other forms of academic dishonesty. Any act of academic dishonesty will subject the student to penalty, including the high probability of failure of the course (i.e., assignment of a grade of “F”). It is expected that you will behave in an honorable and respectful way as you learn and share ideas. Therefore, recycled papers, work submitted to other courses, and major assistance in preparation of assignments without identifying and acknowledging such assistance are not acceptable. All work for this course must be original for this course. Additionally, you are not allowed to post course homeworks, exams, solutions, etc., on a public forum. Please be familiar with the University and the School policies regarding plagiarism. Read the Academic Integrity Policy and Procedure for more information: http://undergrad-catalog.buffalo.edu/policies/course/integrity.shtml. Visit the Senior Vice Provost for Academic Affairs web page for the latest information at http://vpue.buffalo.edu/policies/
Machine Learning Honor Code
Against the ML honor code to:
- Collaborate on Gradiance quizzes
- Collaborate or cheat during exams
- Submit someone else’s work, including from the internet, as one’s own for any submission
- Misuse Piazza forum
You are allowed to:
- Have discussions about homeworks. Every student should submit own homework with names of students in the discussion group explicitly mentioned.
Warning
* Violation of ML honor code and departmental policy will result in an automatic F for the concerned submission * Two violations ⇒ fail grade in the course